Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    phrases indicating levels of difficulty or improbability

    oai_token-act-pair · gpt-3.5-turbo

    the phrase "let alone."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    natureconservancy
    -0.73
    assian
    -0.72
    cill
    -0.69
    idian
    -0.59
    antz
    -0.58
    ets
    -0.56
    edge
    -0.54
    cale
    -0.54
     encount
    -0.54
    peria
    -0.53
    POSITIVE LOGITS
     alone
    1.20
    tered
    0.92
    tering
    0.88
    hetically
    0.85
    ting
    0.80
    icia
    0.77
    ingly
    0.76
    ogether
    0.72
    ardless
    0.70
    ãĥĥãĥī
    0.67
    Act Density 0.021%

    No Known Activations